CN110208771B - Point cloud intensity correction method of mobile two-dimensional laser radar - Google Patents
Point cloud intensity correction method of mobile two-dimensional laser radar Download PDFInfo
- Publication number
- CN110208771B CN110208771B CN201910585835.0A CN201910585835A CN110208771B CN 110208771 B CN110208771 B CN 110208771B CN 201910585835 A CN201910585835 A CN 201910585835A CN 110208771 B CN110208771 B CN 110208771B
- Authority
- CN
- China
- Prior art keywords
- intensity
- laser
- distance
- laser radar
- scanning
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/48—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S17/00
- G01S7/4802—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S17/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Optical Radar Systems And Details Thereof (AREA)
Abstract
A point cloud intensity correction method of ground moving two-dimensional laser starts from factors influencing intensity difference, and according to the principle that the reflection intensities of similar ground objects are the same, intensity data of a standard diffuse reflection plate under different scanning distances and angles are extracted by a two-dimensional laser radar, distance and incidence angle correction models based on data driving are respectively established, and correction parameters are solved by a least square method; then, a mobile two-dimensional laser scanning system is adopted to carry out mobile scanning on the actual scene, information such as the point cloud distance, the scanning angle and the intensity of the actual scene is obtained, and the influence of the distance and the incidence angle on the intensity value is eliminated according to the established correction model, so that the intensity of the corrected homogeneous area tends to be uniform.
Description
Technical Field
The invention relates to a method for correcting the point cloud intensity of a laser, in particular to a method for correcting the point cloud intensity of a ground moving two-dimensional laser radar.
Background
The laser intensity is characterized by the optical power of the scanned target to the backscattered echo of the emitted laser beam. Due to factors such as scanner characteristics, atmospheric transmission, detector and amplifier circuit noise, and geometry of the scanned object, the laser intensity value cannot be directly extracted as a reflection characteristic, but needs to be corrected to eliminate interference of various factors. Through laser intensity correction, the attribute information of the scanned target is fully mined, and research work of classifying by using target characteristics can be carried out.
The current research shows that the corrected laser has important value in the fields of forest and earth surface resource exploration, historical cultural relic restoration, pavement and tunnel damage detection, glacier surface classification and the like. The laser intensity information is often used to assist in improving the ground object point cloud classification accuracy. Aiming at different scanning modes, the existing laser intensity correction methods are mainly divided into the following two types:
(1) Ground three-dimensional laser scanning (TLS) intensity correction. The ground three-dimensional laser scanning can acquire high-precision three-dimensional space information and record the laser intensity of a target. Because the ground three-dimensional laser radar has a relatively short scanning distance and environmental factors such as atmosphere and the like can be ignored, the scanning distance and the incidence angle are mainly considered for laser intensity correction. By fitting the measured data, the relation between the laser intensity value and the distance and incidence angle can be obtained. Meanwhile, the target reflectivity is obtained through the reverse calculation.
(2) Airborne Laser Scanning (ALS) intensity correction. The airborne laser scanning is to obtain high-precision three-dimensional point cloud data by depending on a global positioning system, an inertia measuring device and a laser scanner, and corrects factors such as laser propagation distance, atmospheric attenuation and incidence angle from the attenuation mechanism of laser intensity in the propagation process, so as to eliminate the influence of interference factors on the laser intensity value.
ALS and TLS both adopt three-dimensional laser radars with higher cost, and the mobile two-dimensional laser radars have low cost and small data redundancy, and can directly index point clouds according to the frame number and the intra-frame number of a measuring point. At present, no method is available for eliminating the influence of the distance and the incident angle on the intensity of the mobile two-dimensional lidar, so that an intensive research needs to be carried out on the intensity correction method of the mobile two-dimensional lidar.
Disclosure of Invention
The invention aims to provide a point cloud intensity correction method of ground moving two-dimensional laser aiming at the point cloud intensity difference of a homogeneous region obtained by a moving two-dimensional laser radar.
The technical scheme of the invention is as follows:
the invention provides a point cloud intensity correction method of a mobile two-dimensional laser radar, which comprises the following steps:
s1, extracting intensity data of a standard diffuse reflection plate under different scanning distances and incidence angles by using a two-dimensional laser radar, and respectively establishing functions of the scanning distances and the intensity values and functions of the incidence angles and the intensity values;
s2, resolving correction parameters by adopting a least square method, and determining the order of the model according to the goodness of fit and the root mean square error;
s3, performing mobile scanning on the actual scene by adopting a mobile two-dimensional laser scanning system, acquiring a point cloud distance rho and a scanning angle theta of the actual scene, and calculating an incident angle alpha from a laser foot point to a laser original point;
s4, establishing a correction model of the laser intensity data based on the scanning distance and the incidence angle to obtain the corrected laser intensity I S 。
Further, step S1 specifically includes:
s1-1, setting incidence angle and laser radar distance range [ R ] min ，R max ]And the range interval xm of the lidar; obtaining the laser intensity data of the standard diffuse reflection plate in the distance range by adopting a laser radar at the same incident angle, adjusting the laser intensity data of the standard diffuse reflection plate at each distance received at fixed intervals, and establishing a fitting function f of the scanning distance and the intensity value 2 (R)；
Wherein: r represents the distance from the laser radar to the standard diffuse reflection plate; r is sp Representing the segment points of the fitting function, selecting the distance value at the maximum value of the laser intensity at each distance, wherein n and m respectively represent f 21 (R) and f 22 Polynomial order of (R) [ a ] 0 ，a 1 ，…a n ]And [ b 0 ，b 1 ，…b m ]Coefficients respectively representing polynomials;
s1-2, setting the distance of the laser radar and the incidence angle range of the laser radar (0 degrees and 90 degrees)]And the rotation interval of the standard diffuse reflector; obtaining the laser intensity data of the standard diffuse reflection plate in the incident angle range by adopting a laser radar at the same scanning distance, adjusting the angle according to a fixed rotating interval, receiving at each incident angle, and establishing a fitting function f of the incident angle and the intensity value 3 (cosα)：
f 3 (cosα)＝c 0 +c 1 cosα+…+c k (cosα) k
Wherein: α represents an incident angle of the laser radar; [ c ] A 0 ，c 1 ，…c k ]Denotes f 3 (cos α) polynomial coefficient, k denotes polynomial order.
Further, in step S1-1: for the laser intensity data acquired by the laser radar, taking the middle point of a frame of data as an intensity value at the current distance; in step S1-2: and for the laser intensity data acquired by the laser radar, taking the middle point of one frame of data as the intensity value under the current incident angle.
Further, in step S1-2: the standard diffusely reflecting plate has a rotation interval of 10 °.
Further, step 2 specifically comprises:
s2-1, obtaining a polynomial coefficient [ a ] in the two functions through least square calculation according to the function of the scanning distance and the intensity value established in the step 1 and the function of the incidence angle and the intensity value 0 ，a 1 ，a 2 ，…，a n ] T 、[b 0 ，b 1 ，b 2 ，…，b m ] T 、[c 0 ，c 1 ，c 2 ，…，c k ] T ；
S2-2, calculating a Root Mean Square Error (RMSE) and a goodness of fit (R-square) by adopting the following formulas, and obtaining model orders in the two functions:
wherein: RMSE 1 Representing distance-intensity data points (R) participating in the fitting (i) ，I 2 (i) ) And fitting function f 2 (R (i) ) Root mean square error of (d); i represents the number of distance-intensity data points; m represents the total number of distance-intensity data points; f. of 2 (R (i) ) Represents the fitted value of the scan distance and intensity function for the data point numbered I, I 2 (i) Representing the original intensity value recorded by the laser radar at the distance corresponding to the data point with the number i;
RMSE 2 data points (α) representing incident angle-intensity participating in the fitting j ，I 3 (j) ) And fitting function f 3 (cosα (j) ) Root mean square error of (d); j represents the number of incident angle-intensity data points, N represents the total number of incident angle-intensity data points; α represents a laser incident angle; f. of 3 (cosα (j) ) Denotes the fitted value of the angle of incidence and intensity function for the data point numbered j, I 3 (j) Representing the original intensity value recorded by the laser radar under the incident angle corresponding to the data point with the number of j;
R-square 1 denotes f 2 (R (i) ) Goodness of fit, R-square 2 Denotes f 3 (cosα (j) ) Goodness of fit of (2); mu.s 2 Represents the original intensity I 2 (i) Mean value, mu 3 Represents the original intensity I 3 (j) Mean value;
setting the range of polynomial orders n, m and k, wherein n, m and k belong to [0,7], calculating the Root Mean Square Error (RMSE) and the goodness of fit (R-square) corresponding to all coefficients in the preset range, and selecting the order with the goodness of fit close to 1 and the minimum root mean square error.
Further, step 3 specifically comprises:
step 31, taking the initial position of the laser radar as the origin of coordinates O, establishing a rectangular coordinate system O-xyz, wherein the x-axis direction is the movement direction of the laser radar on the sliding table, and the y-axis direction is the scanning depth of the laser radarThe direction of the z axis is the height direction of the scanned target vertical to the ground, and two-dimensional laser radar is used for collecting the polar coordinate data of yz plane in the jth frame
Wherein N is the scanning point number of the laser radar, and a coordinate transformation formula is as follows:
wherein ρ (i ', j ') and θ (i ', j ') are respectively the distance and the scanning angle of the ith laser spot in the jth ' frame of the laser radar, x (i ', j ') represents the coordinate of the ith laser spot in the jth ' frame in the x direction, y (i ', j ') represents the coordinate of the ith laser spot in the jth ' frame in the depth direction, z (i ', j ') represents the coordinate of the ith laser landing foot in the jth frame in the height direction, Δ t represents the scanning period of the laser radar, and v represents the moving speed v of the laser radar;
step 32, calculating a normal vector of the point cloud, establishing a neighborhood set for each point P (i ', j') in the point cloud through a nearest neighbor algorithm, namely acquiring k nearest neighbor points, and fitting a local plane in the least square sense for the points, namely for each point q in the neighborhood r Establishing a covariance matrix M:
wherein: k is the number of the nearest neighbor, r is the number of the nearest neighbor,
is the three-dimensional centroid through the nearest neighbor;
calculating a normal vector by adopting the following formula;
wherein: l is the number of the characteristic value and takes the values of 1, 2 and 3, lambda L Is the value of the characteristic at the L th,
a characteristic vector corresponding to the L-th characteristic value; through principal component analysis, namely, eigenvalue decomposition is carried out on the covariance matrix M, and the minimum eigenvalue lambda in the M matrix is obtained min The corresponding feature vector is taken as the normal vector of the fitting plane, and is referred to as
Step 33, obtaining the distance R and the incident angle α from the laser landing point P (i ', j') ((x (i ', j'), y (i ', j'), z (i ', j')) to the laser origin O (x (i ', j'), 0) by using the following formula;
in the formula (I), the compound is shown in the specification,
which is indicative of the vector of the incident laser light,
a normal vector is represented.
Further, step 4 specifically includes: the distance correction model is established as follows:
establishing a correction model to obtain the laser intensity I after the distance and the incidence angle are corrected S ：
Wherein; i' denotes the distance which is not corrected, i.e. is to be correctedCorresponding original intensity value, R m Indicating the standard distance, alpha, to be corrected m Indicating the standard angle of incidence, I, to be corrected SR The laser intensity value after distance correction is shown, R represents the distance from the laser foot point P to the laser origin O, and alpha represents the incident angle.
The invention has the beneficial effects that:
the invention relates to a point cloud intensity correction method of mobile two-dimensional laser, which starts from factors influencing intensity difference and utilizes a two-dimensional laser radar to extract intensity data of a standard diffuse reflection plate under different scanning distances and angles according to the principle that the reflection intensities of similar ground objects are the same; then, a mobile two-dimensional laser scanning system is adopted to carry out mobile scanning on the actual scene, information such as the point cloud distance, the scanning angle and the intensity of the actual scene is obtained, and the influence of the distance and the incidence angle on the intensity value is eliminated according to the correction model, so that the intensity of the corrected homogeneous region tends to be uniform.
Additional features and advantages of the invention will be set forth in the detailed description which follows.
Drawings
The above and other objects, features and advantages of the present invention will become more apparent by describing in detail exemplary embodiments thereof with reference to the attached drawings, wherein like reference numerals generally represent like parts in the exemplary embodiments of the present invention.
FIG. 1 is a two-dimensional lidar data acquisition coordinate system
FIG. 2 is a schematic diagram of scanning time and geometry
FIG. 3 is a graph of distance versus intensity
FIG. 4 is a graph of incident angle versus intensity
FIG. 5 is a graph of corrected intensity at different distances
FIG. 6 is a graph of intensity after correction for different angles of incidence
FIG. 7 is a schematic view of a scanned wall surface
FIG. 8 is a pseudo-color image and an intensity distribution histogram after intensity correction of the point cloud of the scanned wall surface.
Fig. 8 (a) is a raw intensity pseudo-color image, fig. 8 (b) is a raw intensity distribution histogram, fig. 8 (c) is an intensity pseudo-color image after an incident angle correction, fig. 8 (d) is an incident angle correction intensity distribution histogram, fig. 8 (e) is an intensity pseudo-color image after a distance correction, and fig. 8 (f) is a distance correction intensity distribution histogram.
Detailed Description
Preferred embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. While the preferred embodiments of the present invention are shown in the drawings, it should be understood that the present invention may be embodied in various forms and should not be limited to the embodiments set forth herein.
The invention provides a point cloud intensity correction method of a mobile two-dimensional laser radar, which comprises the following steps:
s1, extracting intensity data of a standard diffuse reflection plate under different scanning distances and incidence angles by using a two-dimensional laser radar, and respectively establishing functions of the scanning distances and the intensity values and functions of the incidence angles and the intensity values, wherein the functions specifically comprise:
s1-1, setting an incident angle and a laser radar distance range [ R min ，R max ]And the range interval xm of the lidar; obtaining the laser intensity data of the standard diffuse reflection plate in the distance range by adopting a laser radar at the same incident angle, adjusting the laser intensity data of the standard diffuse reflection plate at each distance received at fixed intervals, and establishing a fitting function f of the scanning distance and the intensity value 2 (R)；
Wherein: r represents the distance from the laser radar to the standard diffuse reflection plate; r sp Representing the segment points of the fitting function, selecting the distance value at the maximum value of the laser intensity at each distance, and n and m respectively represent f 21 (R) and f 22 Polynomial order of (R) [ a ] 0 ，a 1 ，…a n ]And [ b 0 ，b 1 ，…b m ]Coefficients respectively representing polynomials;
s1-2, setting the distance of a laser radar, the incidence angle range [0 degrees and 90 degrees ] of the laser radar and the rotation interval (the interval is preferably 10 degrees) of a standard diffuse reflection plate;
obtaining the laser intensity data of the standard diffuse reflection plate in the range of the incidence angle by adopting a laser radar at the same scanning distance, adjusting the angle according to the fixed rotation interval, receiving at each incidence angle, and establishing a fitting function f of the incidence angle and the intensity value 3 (cosα)：
f 3 (cosα)＝c 0 +c 1 cosα+…+c k (cosα) k
Wherein: α represents an incident angle of the laser radar; [ c ] A 0 ，c 1 ，…c k ]Denotes f 3 (cos α) polynomial coefficient, k denotes the polynomial order.
Further, in step S1-1: for the laser intensity data acquired by the laser radar, taking the middle point of one frame of data as an intensity value under the current distance; in step S1-2: and for the laser intensity data acquired by the laser radar, taking the middle point of one frame of data as the intensity value under the current incident angle.
S2, resolving correction parameters by adopting a least square method, and determining a model order according to goodness of fit and root mean square error, wherein the method specifically comprises the following steps:
s2-1, obtaining a polynomial coefficient [ a ] in the two functions through least square calculation according to the function of the scanning distance and the intensity value established in the step 1 and the function of the incidence angle and the intensity value 0 ，a 1 ，a 2 ，…，a n ] T 、[b 0 ，b 1 ，b 2 ，…，b m ] T 、[c 0 ，c 1 ，c 2 ，…，c k ] T ；
S2-2, calculating a Root Mean Square Error (RMSE) and a goodness of fit (R-square) by adopting the following formulas, and obtaining model orders in the two functions:
wherein: RMSE 1 Representing distance-intensity data points (R) participating in the fitting (i) ，I 2 (i) ) And fitting function f 2 (R (i) ) Root mean square error of (d); i represents the number of distance-intensity data points; m represents the total number of distance-intensity data points; f. of 2 (R (i) ) Represents the fitted value of the scan distance and intensity function for the data point numbered I, I 2 (i) Representing the original intensity value recorded by the laser radar under the distance corresponding to the data point with the number i;
RMSE 2 data points (α) representing incident angle-intensity involved in the fitting j ，I 3 (j) ) And fitting function f 3 (cosα (j) ) Root mean square error of (d); j represents the number of incident angle-intensity data points, N represents the total number of incident angle-intensity data points; α represents a laser incident angle; f. of 3 (cosα (j) ) Denotes the fitted value of the angle of incidence and intensity function for the data point numbered j, I 3 (j) Representing the original intensity value recorded by the laser radar under the incident angle corresponding to the data point with the number of j;
R-square 1 denotes f 2 (R (i) ) Goodness of fit, R-square 2 Denotes f 3 (cosα (j) ) The goodness of fit of (2); mu.s 2 Represents the original intensity I 2 (i) Mean value, mu 3 Represents the original intensity I 3 (j) Mean value;
setting the range of polynomial orders n, m and k, wherein n, m and k belong to [0,7], calculating Root Mean Square Errors (RMSE) and goodness of fit (R-square) corresponding to all coefficients in the preset range, and selecting the order with the goodness of fit close to 1 and the minimum root mean square error.
S3, performing mobile scanning on the actual scene by adopting a mobile two-dimensional laser scanning system, acquiring a point cloud distance rho and a scanning angle theta of the actual scene, and calculating an incident angle alpha from a laser foot point to a laser original point;
step 31, taking the initial position of the laser radar as the origin of coordinates O, establishing a rectangular coordinate system O-xyz, wherein the x-axis direction is the movement direction of the laser radar on the sliding table, the y-axis direction is the scanning depth direction of the laser radar, the z-axis direction is the height direction of a scanned target perpendicular to the ground, and collecting polar coordinate data of a yz plane in the jth' frame by using a two-dimensional laser radar
Wherein N is the scanning point number of the laser radar, and a coordinate transformation formula is as follows:
wherein ρ (i ', j') and θ (i ', j') are respectively the distance and the scanning angle of the ith 'laser point in the jth' frame of the laser radar, x (i ', j') represents the coordinate of the ith 'laser point in the jth' frame in the x direction, y (i ', j') represents the coordinate of the ith 'laser point in the jth' frame in the depth direction, z (i ', j') represents the coordinate of the ith 'laser landing foot point in the jth' frame in the height direction, Δ t represents the scanning period of the laser radar, and v represents the moving speed v of the laser radar;
step 32, calculating a normal vector of the point cloud, for each point P (i ', j') in the point cloud, establishing a neighborhood set through a nearest neighbor algorithm, namely obtaining k nearest neighbor points, and then fitting a local plane in the least square sense for the points, namely for each point q in the neighborhood r Establishing a covariance matrix M:
wherein: k is the number of the nearest neighbor, r is the number of the nearest neighbor,
is the three-dimensional centroid passing through the closest point;
calculating a normal vector by using the following formula
Wherein: l is the number of the characteristic values, and takes the values of 0, 1 and 2, lambda min The value of the characteristic is the minimum value,
the characteristic vector is the l-th characteristic vector corresponding to the minimum characteristic value; performing principal component analysis, namely performing eigenvalue decomposition on the covariance matrix M, and taking an eigenvector corresponding to the minimum eigenvalue in the matrix M as a normal vector of a plane;
step 33, obtaining the distance R and the incident angle α from the laser landing point P (i ', j') ((x (i ', j'), y (i ', j'), z (i ', j')) to the laser origin O (x (i ', j'), 0) by using the following formula;
in the formula (I), the compound is shown in the specification,
which represents the vector of the incident laser light,
a normal vector is represented.
S4, establishing a correction model of the laser intensity data based on the scanning distance and the incident angle to obtain the corrected laser intensity I S ；
Wherein; i' represents the original intensity value corresponding to the distance that is not corrected, i.e. to be corrected, R m Indicating the standard distance, alpha, to be corrected m Indicating the standard angle of incidence, I, to be corrected S The laser intensity value after distance correction is shown, R represents the distance from the laser foot point P to the laser origin O, and alpha represents the incident angle.
In the specific implementation: the experiment adopts an intensity correcting unit that accurate at uniform velocity synchronization slip table carried two-dimensional laser radar, and the device mainly comprises slip table mobile control unit and laser data acquisition unit, and laser sensor passes through USB serial ports communication with the computer. The selected UTM-30LX two-dimensional laser scanner produced by HOKUYO company has a scanning range of 0 degree and 270 degrees]The maximum measurement distance is 30m, the angular resolution is 0.25 DEG, and the scanning period is 25ms. Acquiring point cloud data of a homogeneous area of a standard diffuse reflection plate by using a laser radar, taking the standard diffuse reflection plate with the reflectivity of 50% as a scanning object, continuously acquiring 10 frames by aiming at the laser radar for a distance experiment, taking intensity data of a middle point of each frame of data and calculating an average value of the intensity data, arranging 18 stations at a short distance, wherein the scanning distance range is 0.1-1.8 m, arranging one station at intervals of 0.1m for scanning the diffuse reflection plate, arranging 21 stations at a far distance, wherein the scanning range is 1.8-14.4 m, and arranging one station at intervals of 0.6m for scanning; the incident angle experiment is respectively completed under three fixed stations, each station is respectively 1.2m, 1.8m and 2.4m away from the diffuse reflection plate, the diffuse reflection plate is also opposite to the laser radar, for each station, a standard diffuse reflection plate is horizontally rotated every 10 degrees by using an angle measuring instrument within the range of 0-80 degrees and is scanned once, 10 frames of data are also continuously collected, and the intensity data of each frame of middle point is obtained and the average value of the intensity data is calculated. In the polynomial fitting process, by root mean squareError (RMSE) and determining a coefficient (R-square) to evaluate the influence of the polynomial order on the fitting result, f 21 (R)、f 22 (R)、f 3 The determination coefficients of the model (cos θ) are 0.9913, 0.9985 and 0.9979 respectively, which are very close to 1, and the root mean square error is controlled to be 10.2, 5.3 and 6.4, so that the optimal model orders are selected to be n =3, m =5 and k =1 respectively. When the correction model is verified, the laser radar is installed at a position 0.15m away from the ground, the uniform-speed synchronous sliding table is designed to move along the direction parallel to the wall surface, the moving speed is 10mm/s, the wall surface at the position 3m away from the ground is selected in an experiment, the number of point clouds on the selected wall surface is 609248, and the distance range of the wall surface to be corrected is [3m,4m]The incident angle is in the range of [0 DEG, 60 DEG ]]Experiments show that the corrected intensity values are distributed around 3000 in a concentrated mode.
The geometric relationship between the wall point cloud obtained at different distances and angles and the mobile two-dimensional laser radar is shown in fig. 2, and the strength change of the wall surface before and after correction obtained by the method disclosed by the invention is shown in fig. 8. (for convenience of display, the point cloud intensity is colorized), the intensity values distributed in different areas of the same wall surface before correction are found to be quite scattered, and the intensity values distributed after correction are more concentrated and present Gaussian distribution.
Having described embodiments of the present invention, the foregoing description is intended to be exemplary, not exhaustive, and not limited to the embodiments disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments.
Claims (7)
1. A point cloud intensity correction method for a mobile two-dimensional laser radar is characterized by comprising the following steps:
s1, extracting intensity data of a standard diffuse reflection plate under different scanning distances and incidence angles by using a two-dimensional laser radar, and respectively establishing functions of the scanning distances and the intensity values and functions of the incidence angles and the intensity values;
s2, resolving correction parameters by adopting a least square method, and determining a model order according to goodness of fit and root mean square error;
s3, performing mobile scanning on the actual scene by adopting a mobile two-dimensional laser scanning system, acquiring a point cloud distance rho and a scanning angle theta of the actual scene, and calculating an incident angle alpha from a laser foot point to a laser original point;
s4, establishing a correction model of the laser intensity data based on the scanning distance and the incidence angle to obtain the corrected laser intensity I S 。
2. The point cloud intensity correction method for a mobile two-dimensional lidar according to claim 1, wherein the step S1 specifically comprises:
s1-1, setting an incident angle and a laser radar distance range [ R min ，R max ]And the distance interval xm of the lidar; obtaining the laser intensity data of the standard diffuse reflection plate in the distance range by adopting a laser radar at the same incident angle, adjusting the laser intensity data of the standard diffuse reflection plate at each distance received at fixed intervals, and establishing a fitting function f of the scanning distance and the intensity value 2 (R)；
Wherein: r represents the distance from the laser radar to the standard diffuse reflection plate; r sp Representing the segment points of the fitting function, selecting the distance value at the maximum value of the laser intensity at each distance, wherein n and m respectively represent f 21 (R) and f 22 (polynomial order of R, [ a ] 0 ，a 1 ，…a n ]And [ b) 0 ，b 1 ，…b m ]Coefficients respectively representing polynomials;
s1-2, setting the distance of the laser radar and the incidence angle range of the laser radar (0 degrees and 90 degrees)]And a rotation interval of the standard diffuse reflector; obtaining the laser intensity data of the standard diffuse reflection plate in the range of the incidence angle by adopting a laser radar at the same scanning distance, adjusting the angle according to the fixed rotation interval, receiving at each incidence angle, and establishing a fitting function f of the incidence angle and the intensity value 3 (cosα)：
f 3 (cosα)＝c 0 +c 1 cosα+…+c k (cosα) k
Wherein: α represents an incident angle of the laser radar; [ c ] A 0 ，c 1 ，…c k ]Denotes f 3 (cos α) polynomial coefficient, k denotes polynomial order.
3. The method for correcting the point cloud intensity of a mobile two-dimensional lidar according to claim 2, wherein in step S1-1: for the laser intensity data acquired by the laser radar, taking the middle point of a frame of data as an intensity value at the current distance; in step S1-2: and for the laser intensity data acquired by the laser radar, taking the middle point of one frame of data as the intensity value under the current incident angle.
4. The method for correcting the point cloud intensity of a mobile two-dimensional lidar according to claim 2, wherein in step S1-2: the standard diffusely reflecting plate has a rotation interval of 10 °.
5. The point cloud intensity correction method for a mobile two-dimensional lidar according to claim 2, wherein the step 2 is specifically:
s2-1, obtaining a polynomial coefficient [ a ] in the two functions through least square calculation according to the function of the scanning distance and the intensity value established in the step 1 and the function of the incidence angle and the intensity value 0 ，a 1 ，a 2 ，…，a n ] T 、[b 0 ，b 1 ，b 2 ，…，b m ] T 、[c 0 ，c 1 ，c 2 ，…，c k ] T ；
S2-2, calculating a Root Mean Square Error (RMSE) and a goodness of fit (R-square) by adopting the following formulas, and obtaining model orders in the two functions:
wherein: RMSE 1 Representing distance-intensity data points (R) participating in the fitting (i) ，I 2 (i) ) And fitting function f 2 (R (i) ) Root mean square error of (d); i represents the number of distance-intensity data points; m represents the total number of distance-intensity data points; f. of 2 (R (i) ) Represents the fitted value of the scan distance and intensity function for the data point numbered I, I 2 (i) Representing the original intensity value recorded by the laser radar at the distance corresponding to the data point with the number i;
RMSE 2 data points (α) representing incident angle-intensity participating in the fitting j ，I 3 (j) ) And fitting function f 3 (cosα (j) ) Root mean square error of (d); j represents the number of incident angle-intensity data points, N represents the total number of incident angle-intensity data points; α represents a laser incident angle; f. of 3 (cosα (j) ) Denotes the fitted value of the angle of incidence and intensity function for the data point numbered j, I 3 (j) Representing the original intensity value recorded by the laser radar under the incident angle corresponding to the data point with the number of j;
R-square 1 denotes f 2 (R (i) ) Goodness of fit, R-square 2 Denotes f 3 (cosα (j) ) Goodness of fit of (2); mu.s 2 Represents the original intensity I 2 (i) Mean value, mu 3 Represents the original intensity I 3 (j) Mean value;
setting the range of polynomial orders n, m and k, wherein n, m and k belong to [0,7], calculating Root Mean Square Errors (RMSE) and goodness of fit (R-square) corresponding to all coefficients in the preset range, and selecting the order with the goodness of fit close to 1 and the minimum root mean square error.
6. The method for correcting the point cloud intensity of a mobile two-dimensional lidar according to claim 1, wherein step 3 comprises:
step 31, taking the initial position of the laser radar as a coordinate origin O, establishing a rectangular coordinate system O-xyz, taking the x-axis direction as the movement direction of the laser radar on the sliding table, the y-axis direction as the scanning depth direction of the laser radar, and the z-axis direction as the height direction of a scanned target perpendicular to the ground, and collecting polar coordinate data of a yz plane in the jth' frame by using a two-dimensional laser radar
Wherein N is the scanning point number of the laser radar, and a coordinate transformation formula is as follows:
wherein ρ (i ', j ') and θ (i ', j ') are respectively the distance and the scanning angle of the ith laser spot in the jth ' frame of the laser radar, x (i ', j ') represents the coordinate of the ith laser spot in the jth ' frame in the x direction, y (i ', j ') represents the coordinate of the ith laser spot in the jth ' frame in the depth direction, z (i ', j ') represents the coordinate of the ith laser landing foot in the jth frame in the height direction, Δ t represents the scanning period of the laser radar, and v represents the moving speed v of the laser radar;
step 32, calculating a normal vector of the point cloud, establishing a neighborhood set for each point P (i ', j') in the point cloud through a nearest neighbor algorithm, namely acquiring k nearest neighbor points, and fitting a local plane in the least square sense for the points, namely for each point q in the neighborhood r Establishing a covariance matrix M:
wherein: k is the number of the nearest neighbor, r is the number of the nearest neighbor,
is the three-dimensional centroid passing through the closest point;
calculating a normal vector by adopting the following formula;
wherein: l is the number of the characteristic value, and takes values of 1, 2 and 3, lambda L Is the value of the characteristic at the L th,
the characteristic vector corresponding to the L-th characteristic value; through principal component analysis, namely, eigenvalue decomposition is carried out on the covariance matrix M, and the minimum eigenvalue lambda in the M matrix is obtained min The corresponding feature vector is taken as the normal vector of the fitting plane, and is referred to as
Step 33, obtaining the distance R and the incident angle α from the laser landing point P (i ', j') ((x (i ', j'), y (i ', j'), z (i ', j')) to the laser origin O (x (i ', j'), 0) by using the following formula;
in the formula (I), the compound is shown in the specification,
which represents the vector of the incident laser light,
a normal vector is represented.
7. The point cloud intensity correction method for a mobile two-dimensional lidar according to claim 1, wherein step 4 specifically comprises: the distance correction model is established as follows:
establishing a correction model to obtain the laser intensity I after the distance and the incidence angle are corrected S ：
Wherein; i' denotes the original intensity value, R, corresponding to the distance that is not corrected, i.e. to be corrected m Indicating the standard distance, alpha, to be corrected m Indicating the standard angle of incidence, I, to be corrected S The laser intensity value after distance correction is shown, R is the distance from the laser landing point P to the laser origin O, and alpha is the incidence angle.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910585835.0A CN110208771B (en) | 2019-07-01 | 2019-07-01 | Point cloud intensity correction method of mobile two-dimensional laser radar |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910585835.0A CN110208771B (en) | 2019-07-01 | 2019-07-01 | Point cloud intensity correction method of mobile two-dimensional laser radar |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110208771A CN110208771A (en) | 2019-09-06 |
| CN110208771B true CN110208771B (en) | 2022-12-30 |
Family
ID=67795689
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910585835.0A Active CN110208771B (en) | 2019-07-01 | 2019-07-01 | Point cloud intensity correction method of mobile two-dimensional laser radar |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110208771B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111781611B (en) * | 2020-06-16 | 2023-06-16 | 万物镜像(北京)计算机系统有限公司 | Method and device for establishing model, storage medium and electronic equipment |
| CN111679262B (en) * | 2020-07-06 | 2023-09-01 | 武汉海达数云技术有限公司 | Laser point cloud intensity calibration method, device, equipment and storage medium |
| CN112526534B (en) * | 2020-11-03 | 2024-03-08 | 上海炬佑智能科技有限公司 | ToF sensing device and distance detection method thereof |
| CN112526535B (en) * | 2020-11-03 | 2024-03-08 | 上海炬佑智能科技有限公司 | ToF sensing device and distance detection method thereof |
| CN113033564B (en) * | 2021-02-20 | 2022-10-14 | 意欧斯物流科技(上海)有限公司 | High-precision and robust 2D laser point cloud matching method |
| CN114322828A (en) * | 2021-12-15 | 2022-04-12 | 云南昆钢电子信息科技有限公司 | Point cloud sampling system and method based on laser radar scanning angle switching |
| CN115166757B (en) * | 2022-07-25 | 2023-08-01 | 珠海市睿特光电技术有限公司 | Method, system and storage medium for measuring actual detection distance of laser radar |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2012020696A1 (en) * | 2010-08-11 | 2012-02-16 | 株式会社トプコン | Device for processing point group position data, system for processing point group position data, method for processing point group position data and program for processing point group position data |
| CN106097423A (en) * | 2016-06-08 | 2016-11-09 | 河海大学 | LiDAR point cloud intensity correction method based on k neighbour |
| CN107179533A (en) * | 2017-05-03 | 2017-09-19 | 长安大学 | A kind of airborne LiDAR systematic errors Self-checking method of multi-parameter |
| CN108830921A (en) * | 2018-05-16 | 2018-11-16 | 陈年康 | Laser point cloud reflected intensity correcting method based on incident angle |
| CN109725303A (en) * | 2018-12-04 | 2019-05-07 | 北京万集科技股份有限公司 | Modification method and device, the storage medium of coordinate system |
-
2019
- 2019-07-01 CN CN201910585835.0A patent/CN110208771B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2012020696A1 (en) * | 2010-08-11 | 2012-02-16 | 株式会社トプコン | Device for processing point group position data, system for processing point group position data, method for processing point group position data and program for processing point group position data |
| CN106097423A (en) * | 2016-06-08 | 2016-11-09 | 河海大学 | LiDAR point cloud intensity correction method based on k neighbour |
| CN107179533A (en) * | 2017-05-03 | 2017-09-19 | 长安大学 | A kind of airborne LiDAR systematic errors Self-checking method of multi-parameter |
| CN108830921A (en) * | 2018-05-16 | 2018-11-16 | 陈年康 | Laser point cloud reflected intensity correcting method based on incident angle |
| CN109725303A (en) * | 2018-12-04 | 2019-05-07 | 北京万集科技股份有限公司 | Modification method and device, the storage medium of coordinate system |
Non-Patent Citations (2)
| Title |
|---|
| 基于激光雷达强度值的目标反射特征提取;童祎 等;《激光与光电子学进展》;20180511(第10期);全文 * |
| 激光入射角度对点云反射强度的影响研究;夏国芳 等;《激光杂志》;20160425;第37卷(第04期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110208771A (en) | 2019-09-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110208771B (en) | Point cloud intensity correction method of mobile two-dimensional laser radar | |
| US7436522B2 (en) | Method for determining the 3D coordinates of the surface of an object | |
| CN102607457B (en) | Measuring device and measuring method for large three-dimensional morphology based on inertial navigation technology | |
| US20210056716A1 (en) | Combined point cloud generation using a stationary laser scanner and a mobile scanner | |
| Shang et al. | Measurement methods of 3D shape of large-scale complex surfaces based on computer vision: A review | |
| CN105203023A (en) | One-stop calibration method for arrangement parameters of vehicle-mounted three-dimensional laser scanning system | |
| CN110415259A (en) | A kind of shade tree point cloud recognition methods based on laser reflection intensity | |
| CN110966956A (en) | Binocular vision-based three-dimensional detection device and method | |
| CN115164724A (en) | Shafting error model of ground three-dimensional laser scanner and calibration method | |
| CN114323571A (en) | Multi-optical-axis consistency detection method for photoelectric aiming system | |
| CN106546766B (en) | Clinoplain scan method based on two anemometry laser radars | |
| Crispel et al. | All-sky photogrammetry techniques to georeference a cloud field | |
| Ullrich et al. | Long-range high-performance time-of-flight-based 3D imaging sensors | |
| CN112581524A (en) | Point cloud-based SLAM mobile robot airport road detection data acquisition method | |
| CN115902816A (en) | Automatic measuring system for engineering measurement | |
| CN115451846B (en) | TLS-based large parabolic antenna caliber measuring and calculating method | |
| CN114663486A (en) | Building height measurement method and system based on binocular vision | |
| Tu et al. | Geometrical model of laser triangulation system based on synchronized scanners | |
| CN115371599A (en) | High-precision ground flatness measuring system and method | |
| US20210233276A1 (en) | Imaging system | |
| CN113888693A (en) | High-precision point cloud data reconstruction method | |
| Wang et al. | Precision estimation of mobile laser scanning system | |
| Hlotov et al. | Accuracy investigation of point clouds with Faro Focus 3d S120 terrestrial laser scanner | |
| Shan et al. | A laser triangulation based on 3d scanner used for an autonomous interior finishing robot | |
| CN117029689B (en) | Laser triangulation system and tunnel scanning method using same |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |